MOLEK-SYNTEZ by Zachtronics

Looks a lot like Opus Magnum (hex grid, atoms forming bonds, similar progamming model). But the things you use to manipulate the molecules appear to be restricted to the perimeter of the board.

Wishlisted for 1.0 release.

Nothing that Zachtronics has made has worked for me as well as SpaceChem (not even Opus Magnum–I blame the hexes), but I’m happy to see them appear anyway. This one is an odd theme, but… why not!

Oh, so like 75% of my playthrough of Opus Magnum before I figured out you could scroll.

I liked the game better before I found out, so I’m in.

Just feels like a mash-up of Opus Magnum and Spacechem. But I’m in. ;)

I’m pretty sure you can scroll in Molek as well, once you get past the first molecule.

This looks superficially like Opus Magnum, but ends up feeling pretty different once you get rolling. I like it.

Looks like a sneaky way of making you take Organic Chemistry. I already did that once!

Ok, Molek-Syntez is definitely minor Zachtronics. Basically this is to Opus Magnum what TIS-100 was to Shenzhen IO. Except that in the latter case, the simple version was released first.

The rules of the game (how bonds are formed and removed, in what order do simultaneous events gets resolved) are unstated and sometimes quite arbitrary. Most annoyingly, the bond/creation order depends partially on the absolute board coordinates. Which is garbage, since now if something doesn’t work the way you want it, what you end up doing is rotating the whole solution. The worst part is that I still haven’t figured out exactly how this works; there are posts on Reddit that claim to have the reverse engineered the rules, but they’re obviously not correct all the time.

They’ve lost all the UI niceties from Opus Magnum somewhere along the way. The one that annoys me the most is that as far as I can tell there’s now no way of copy-pasting program fragments. (And hey, remember from the last paragraph how you might need to rotate the whole solution to solve rotation-based order of operation issues. There’s no batch rotate, just a batch translate. Ho, ho, ho.)

Finally, what’s the optimization like? It’s very different form Opus Magnum, since the constraints you have are on the manipulators (always exactly six) and the program size. Which is good, I hated how the unlimited program size both totally trivialized the “use the minimum number of pieces” optimization target in OM, and how sometimes the optimal solution required inputting the program for the whole duration rather than depending on looping.

In OM, the optimal solutions tended to be about pipelining. Figuring out how to do each part of the solution in exactly four cycles, and handing the pieces off to another manipulator, and then starting its own part of the solution . Molek-Syntez is nothing like that. It’s more about having all the manipulators work on the same single solution simultaneously, and having it all be done in a few cycles.

But the way the last two paragraphs combine is that to do cycle count optimization (the only one I really care about), you basically have to do multiple solutions. Is there a 4 manipulator 2 cycle solution?`Well, that leaves 2 manipulators unused. So you’ll probably try to build a 4 cycle solution with them, and have both of these totally independent solutions work in parallel 33% speedup.

What if the optimal 2 manipulator solution is 5 cycles? Well, then you need to copy the program for that twice, and the copy program for the 4 manipulator part 5 times to make both of them have the same length. Which is just not anywhere near as interesting as trying to get the timing of a single integrated clockwork solution to line up. (Oh, and remember my complaint about how you can’t copy-paste programs? This is why it’s so annjoying. Have fun manually re-entering 14 copies of the 2 instruction programs for each of 5 manipulators, suckers.)

So I’d not recommend this until we see what they can do to fix up the UI during the early access period. Bu the rule set also seem kind of ill-conceived, given what it implies for the optimization puzzle. It might not be salvageable at all.

Well, it’s out of EA. I don’t know what was changed, if anything.

Basically nothing, as far as I can tell. They added a bonus campaign and fixed an obscure bug when using a specific graphics option. But no quality of life improvements in the UI, and no attempt to fix the rules to create more interesting puzzles.

That was a really disappointing EA period.